The R2R3-MYB transcription factor FaMYB63 participates in regulation of eugenol production in strawberry

Abstract

The biosynthetic pathway of volatile phenylpropanoids, including 4-allyl-2-methoxyphenol (eugenol), has been investigated in petunia (Petunia hybrida). However, the regulatory network for eugenol accumulation in strawberry (Fragaria × ananassa Duch.) fruit remains unclear. Here, an R2R3-type MYB transcription factor (TF; FaMYB63) was isolated from strawberry by yeast one-hybrid (Y1H) screening using the promoter of the FaEGS1 (eugenol synthase 1 [EGS 1]) gene, which encodes the enzyme responsible for the last step in eugenol biosynthesis. FaMYB63 is phylogenetically distinct from other R2R3-MYB TFs, including FaEOBІІ (EMISSION OF BENZENOID II [EOBII]), which also participates in regulating eugenol biosynthesis in strawberry receptacles. Reverse transcription quantitative PCR (RT-qPCR) assays showed that the expression of FaMYB63 was tissue-specific and consistent with eugenol content through strawberry fruit development, was repressed by abscisic acid, and was activated by auxins (indole-3-acetic acid). Overexpression and RNA interference-mediated silencing of FaMYB63 resulted in marked changes in the transcript levels of the biosynthetic genes FaEGS1, FaEGS2, and FaCAD1 (cinnamyl alcohol dehydrogenase 1 [CAD1]) and, thereby, the accumulation of eugenol. Electrophoretic mobility shift, Y1H, GUS activity, and dual-luciferase activity assays demonstrated that the transcript levels of FaEOBІІ and FaMYB10 were regulated by FaMYB63, but not the other way around. Together, these results demonstrate that FaMYB63 directly activates FaEGS1, FaEGS2, FaCAD1, FaEOBІІ, and FaMYB10 to induce eugenol biosynthesis during strawberry fruit development. These findings deepen the understanding of the regulatory network that influences eugenol metabolism in an edible fruit crop.

Introduction

Strawberry (Fragaria × ananassa Duch.) is an important commodity worldwide due to its nutritional value and unique flavor. The aroma of cultivated strawberries is an important organoleptic attribute that influences their acceptability to plant breeders, the food industry, and consumers. The flavor components of strawberries have been studied extensively, with more than 360 volatile compounds reported (Zorrilla-Fontanesi et al., 2012; Ulrich et al., 2018), including esters, furanones, sulfur compounds, lactones, alcohols, and carbonyls (Pyysalo et al., 1979; Zabetakis and Holden, 1997). However, only 15–20 of these volatiles are key odorants in wild strawberry (Fragaria vesca) accessions. Due to genetic erosion and human selection, the commercially grown modern octoploid strawberry varieties (like F. × ananassa) are devoid of some important odor-active compounds, such as 4-allyl-2-methoxyphenol (eugenol), a phenylpropanoid derivative. The contents of eugenol and isoeugenol in octoploid strawberry are substantially lower than those in diploid wild strawberry, where they contribute to a unique flavor (Pyysalo et al., 1979; Larsen and Poll, 1992; Ulrich et al., 2007; Hoffmann et al., 2011; Aragüez et al., 2013).

Eugenol is a naturally occurring phenylpropanoid and accumulates to high levels in various cultivated plants, such as berries, basil (Ocimum basilicum), cinnamon (Cinnamomum cassia), cloves (Syzygium aromaticum), lemon balm (Melissa officinalis), and nuts, with a variety of biological effects and health benefits (Gang et al., 2002; Koeduka et al., 2006; Pavithra, 2014). In the plant, eugenol acts as a floral attractant for pollinators and as a defense compound against animals and microorganisms (Koeduka et al., 2006; Pasay et al., 2010). Eugenol has received increasing attention due to its food preservative and repellant activities as well as its anti-inflammatory, antiviral, antimicrobial, and pharmacological properties (Sun et al., 2016; Barboza et al., 2018). A clinical study showed that eugenol has a dual mode of action in combating diabetes by lowering blood glucose (Singh et al., 2016).

To date, a proposed biosynthesis pathway of the volatile phenylpropanoids has been illustrated, and the key genes encoding the related enzymes have been characterized (Spitzer-Rimon et al., 2012; Medina-Puche et al., 2015). The precursor of eugenol synthesis is L-phenylalanine, which contains carbon skeletons from the shikimate pathway and is converted through subsequent enzymatic reactions, for example, phenylalanine ammonia lyase (PAL), cinnamate-4-hydroxylase (C4H), p-coumarate 3-hydroxylase (C3H), caffeic acid 3-O-methyltransferase (COMT), 4-coumarate-CoA ligase (4CL), cinnamoyl-CoA reductase (CCR), cinnamyl alcohol dehydrogenase (CAD), coniferyl alcohol acetyltransferase (CFAT), and EGS (Figure 1). In strawberry, FaEGS1, FaEGS2, and FaCAD1 play essential roles in eugenol production (Aragüez et al., 2013; Medina-Puche et al., 2015). FaCAD1 is involved in the reduction of cinnamaldehydes into cinnamyl alcohols (Blanco-Portales et al., 2002). Afterward, the FaEGSs synthesize eugenol from coniferyl acetate (Aragüez et al., 2013). The genes encoding CAD have been identified in different angiosperms and gymnosperms, such as in the complete genome sequencing and annotation of rice (Oryza sativa) and Arabidopsis (Arabidopsis thaliana; Sasaki and Sederoff, 2003; Deng et al., 2013). EGS genes have been isolated from and functionally characterized in basil (O. basilicum; Koeduka et al., 2006), Clarkia breweri, petunia (Petunia hybrida; Koeduka et al., 2006), rose (Rosa chinensis; Wang et al., 2012; Yan et al., 2012), strawberry (F. × ananassa; Aragüez et al., 2013), creosote bush (Larrea tridentata; Vassão et al., 2007), and anise (Pimpinella anisum; Koeduka et al., 2009). In strawberry, FaEGS1 is mainly expressed in green achenes and receptacles, whereas the transcript level of FaEGS2 is predominant in the red-ripening receptacles (Aragüez et al., 2013; Molina-Hidalgo et al., 2017). However, the regulatory network that controls the metabolic pathway leading to the volatile phenylpropanoids, like coniferyl alcohol and coniferyl acetate, that can be converted to eugenol is still not completely understood.

Figure 1

Biosynthetic pathway of eugenol. CHS, chalcone synthase; Blue box: TF, Black arrows: previously known pathway components; red arrows represent results shown in this study.

Transcription factors (TFs) modulate plant development and responses to biotic and abiotic stresses. Transcriptional regulation of phenylpropanoid metabolism has been extensively studied in petunia, and it has been determined that eugenol metabolism is regulated by multiple R2R3-MYB TFs, such as ODORANT1 (ODO1), EMISSION OF BENZENOIDS I (EOBI), EOBII, and PhMYB4 (Spitzer-Rimon et al., 2010, 2012; Colquhoun et al., 2011). Among these genes, only one homolog, FaEOBII, has been identified and functionally characterized in strawberry. FaEOBII has been shown to regulate eugenol biosynthesis by activating the expression of FaCAD1 and FaEGS2 in strawberry (Medina-Puche et al., 2015). Additionally, the expression of FaEOBII has been demonstrated to be further activated by FaMYB10, another R2R3-MYB TF that regulates early and late biosynthetic genes in the flavonoid/phenylpropanoid pathway (Medina-Puche et al., 2014, 2015; Wei et al., 2018). A recent study showed that a TF from the DNA-binding One Zinc Finger (DOF) family (FaDOF2) interacts with FaEOBII and is involved in eugenol production through transactivation of FaEGS2 in ripe fruit receptacles and in petals. However, the expression of FaDOF2 is not under the control of FaMYB10 (Molina-Hidalgo et al., 2017). Therefore, the regulatory roles of the different components in this phenylpropanoid metabolism network are multi-layered and require further exploration.

Most of the identified regulators of the phenylpropanoid pathway play prominent roles in the maturation and ripening processes in the later stages of strawberry fruit development, for example, FaMYB10, FaEOBII, and FaDOF2 (Medina-Puche et al., 2014, 2015; Molina-Hidalgo et al., 2017). However, the identity of the signal(s) involved during the initial developmental stages of strawberry remains largely unknown. Strawberry is a nonclimacteric fruit and ripens without respiration and ethylene, but with chlorophyll degradation and metabolic changes related to starch/sucrose, flavonoids, and phenylpropanoids (Bai et al., 2021). The strawberry is a pseudocarp that originates from the expansion of the flower base (the receptacle), while the true one-seeded fruits (achenes) are localized on the epidermal layer (Aharoni and O’Connell, 2002; Koeduka et al., 2006). Auxin is synthesized in the achene and transported to the receptacle, where it strongly affects the expression of genes during the initial growth and development of the receptacle and delays the ripening process (Aharoni and O’Connell, 2002; Bai et al., 2021). Abscisic acid (ABA) is predominantly produced in the receptacle and promotes ripening of strawberry in an ethylene-independent manner (Chai et al., 2011; Jia et al., 2011; Gu et al., 2019). The use of a custom-made oligonucleotide-based strawberry microarray platform showed that 1,281 genes were upregulated in red receptacles compared to green receptacles. ABA induces ripening-related genes (Medina-Puche et al., 2016), including biosynthetic genes and TFs involved in different organoleptic properties such as aroma, color, taste, and softening, such as quinone oxidoreductase (FaQR; Raab et al., 2006; Daminato et al., 2013), polygalacturonase (FaPG1; Quesada et al., 2009; Daminato et al., 2013), rhamnogalacturonate lyase 1 (FaRGLyase1; Molina-Hidalgo et al., 2013), nodulin 26-like intrinsic protein (FaNIP1;1; Molina-Hidalgo et al., 2015), β-galactosidase 4 (FaβGal4; Paniagua et al., 2016), SHATTERPROOF-like (FaSHP; Daminato et al., 2013), FaMYB10 (Medina-Puche et al., 2014), FaEOBII (Medina-Puche et al., 2015), FaDOF2 (Molina-Hidalgo et al., 2017), and PACLOBUTRAZOL RESISTANCE 1 (FaPRE1; Medina-Puche et al., 2019). In contrast, auxins negatively regulated the majority of ripening-related genes. Six out of the 524 downregulated genes whose expression was induced by auxins and repressed by ABA (Medina-Puche et al., 2016) have not been characterized.

There is a high eugenol content in green, unripe receptacles of strawberry fruit (Medina-Puche et al., 2015; Molina-Hidalgo et al., 2017). However, it is unknown how this accumulation of eugenol is regulated, although it is likely through transcriptional control. In this study, we used a yeast one-hybrid (Y1H) assay to find proteins that interact with the promoter of FaEGS1 in strawberry fruit. We further characterized one potential interactor, the R2R3-MYB TF FaMYB63. The expression of FaMYB63 during fruit development was investigated by RT-qPCR. Functional characterization of FaMYB63 was carried out by transient overexpression (OVX) and silencing. We found that FaMYB63 increases eugenol production by regulating genes encoding biosynthetic enzymes (FaEGS1, FaEGS2, and FaCAD1) and two other R2R3-MYB TFs (FaEOBII and FaMYB10) that regulate eugenol biosynthesis (Medina-Puche et al., 2014, 2015). The results of this study provide evidence for modifying the eugenol regulatory network in strawberry fruit.

Results

Identification and subcellular localization of FaMYB63

To identify TFs involved in the regulation of FaEGS1, a Y1H assay using the promoter of FaEGS1 as the bait was used to screen a cDNA library derived from strawberry fruit. Four putative candidates were obtained, and one was annotated as a TF from the MYB family and analyzed for its expression associated with eugenol accumulation across the seven developmental stages. The MYB-like TF showed 50% amino acid similarity with AtMYB63 from A. thaliana (GenBank accession number NP_178039.1), so it was designated as FaMYB63. The full-length cDNA sequence of the FaMYB63 gene (GenBank accession no. MW452942 and strawberry genome accession number FvH4_3g15320.1) contained an open reading frame of 1,134 bp, which encoded a polypeptide of 378 amino acid residues with a conserved R2R3 binding domain near its N-terminus (Supplemental Figure S1). FaMYB63 had a calculated molecular mass of 42.13 kDa. Multiple sequence alignment indicated that FaMYB63 was clustered into subgroup 3 of the R2R3-MYB TFs (Supplemental Figure S2) and might function in the nucleus. To determine the intracellular localization of FaMYB63, a CaMV35S:FaMYB63-GFP construct and a CaMV35S:GFP construct, as a control, were introduced into Nicotiana benthamiana leaves. Microscopy revealed that FaMYB63-GFP was located exclusively in the nucleus and overlapped with the 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) staining, whereas CaMV35S:GFP was distributed evenly throughout the cell (Figure 2A).

Figure 2

FaMYB63 localization and expression and eugenol content in different stages and under different treatments. A, Subcellular localization of CaMV35S:FaMYB63–GFP fusions in transiently transformed N. benthamiana leaves. All experiments were assayed 72 h after infiltration. The same N. benthamiana leaves were stained with DAPI to show the locations of the nuclei. Bars = 20 μm. B, Analysis by RT-qPCR of FaMYB63 transcript levels (left axis) and the eugenol content (right axis) in fruit receptacles at different developmental stages of Fragaria × ananassa cv. ‘Benihoppe’. C, Analysis by RT-qPCR of FaMYB63 transcript levels in achenes at different developmental stages of F. × ananassa cv. ‘Benihoppe’. G1, small green fruit stage; G2, large green fruit stage; G3, green-white fruit stage; W, white fruit stage; T, red-turning fruit stage; R, red-ripening fruit stage; OR, over-ripening fruit stage. Relative expression values were relative to receptacles at the G1 stage in all cases, which was assigned an arbitrary value equal to one. In (D) and (E), achenes were removed from developing fruits in the G2 stage of F. × ananassa cv. ‘Benihoppe’ for analysis of (D) FaMYB63 expression by RT-qPCR (left axis) and eugenol content (right axis) and (E) IAA content. Control + G2, control, large green fruit receptacle; G2-A + L, G2 fruit receptacle without achenes for 5 d covered by a lanolin paste; G2-A + IAA + L, G2 fruit receptacle without achenes for 5 d treated with the synthetic auxin IAA (1 mM) in lanoline paste. F and G, Analysis by RT-qPCR of FaMYB63 and FaNCED1 transcript levels (in bars) and content of eugenol (in lines) and (H and I) ABA content. F and H, Control, green-white fruits were injected with sterile water; NDGA, green-white fruits were injected with 100-μM NDGA. G and I, Control, fruit pedicels were immersed in MS medium containing sucrose; Water Stress, fruit pedicels were exposed to air without MS medium. Values are the mean ± sd of three biological replicates. The different letters above the columns in (B) and (C) indicate significant differences at the 5% level (P < 0.05, Duncan’s test). Statistical significance with respect to the reference sample (control) was determined by the Student’s t test: *P < 0.05, **P < 0.01, ***P < 0.001.

Expression patterns of FaMYB63

The expression of FaMYB63 was tracked through the seven stages of strawberry fruit development by RT-qPCR (Figure 2B). In vegetative tissues, such as stolons and leaves, FaMYB63 transcripts were high (Supplemental Figure S3A). In receptacles, FaMYB63 transcript levels were the highest at the small green fruit stage (G1) and were second highest in over-ripening (OR) fruit. The lowest value in the receptacles was at the white fruit stage (W). FaMYB63 transcript levels correlated with eugenol content in the receptacles (Figure 2B). In achenes, FaMYB63 was at distinctly high expression levels, even higher than in the receptacles. In achenes, FaMYB63 transcript levels also peaked at the small green fruit stage, followed by the red-turning fruit stage (T), and were significantly lower at the red-ripening (R) and OR stages (Figure 2C). Moreover, the eugenol content in achenes was much higher than that in receptacles (Supplemental Figure S3B). These results indicated that FaMYB63 expression had tissue- and stage-specific profiles.

In strawberry, a nonclimacteric fruit, ABA and auxin are important regulators that act together or independently in response to plant growth and development and act as environmental cues (Perkins-Veazie, 1995; Bai et al., 2021). Auxin produced in achenes is involved in receptacle development and subsequently suppresses fruit ripening (Gu et al., 2019). ABA biosynthesis in receptacles plays a vital role in fruit ripening (Jia et al., 2011; Estrada-Johnson et al., 2017). To ascertain whether the phytohormone indole-3-acetic acid (IAA) regulates FaMYB63 expression, de-achened fruits at the large green stage （G2）were treated with or without IAA in lanolin paste according to Medina-Puche et al. (2015). The expression of FaMYB63 was significantly decreased with removal of the achenes compared with the control fruits. The external application of IAA enhanced FaMYB63 transcript levels (Figure 2D). The eugenol content correlated with the IAA content (Figure 2E). These data suggested that FaMYB63 expression was under the positive control of auxin.

To evaluate whether FaMYB63 expression is responsive to ABA treatment, two different experiments were conducted according to Medina-Puche et al. (2015): (1) injection into the interior of the fruit receptacles of nordihydroguaiaretic acid (NDGA), an inhibitor of 9-cis-epoxycarotenoid dioxygenase activity, the limiting enzyme in the ABA biosynthetic pathway and (2) withholding water from plants, which results in accumulation of ABA in plants (Huang et al., 2008). As shown in Figure 2, F–I, the ABA content and FaNCED1 transcript level showed opposite patterns to the eugenol content and FaMYB63 transcript level. These results indicated that ABA might significantly downregulate the expression of FaMYB63 and added insights into ABA regulation in strawberry when combined with previous studies (Jia et al., 2011; Medina-Puche et al., 2015).

FaMYB63 transactivates the expression of genes involved in eugenol production in strawberry fruit receptacles

The potential of FaMYB63 to regulate the expression of genes in the eugenol biosynthesis pathway in strawberry fruits was assessed by RT-qPCR. Fruits transiently expressing FaMYB63 were obtained by injection of Agrobacterium tumefaciens GV3101 carrying the OVX plasmid pCXSN-FLAG-FaMYB63 or the silencing (RNA interference [RNAi]) plasmid pHellsgate2-FaMYB63 and were named FaMYB63-OVX1, FaMYB63-OVX2, and FaMYB63-OVX3 and FaMYB63-RNAiTA, FaMYB63-RNAiTB, and FaMYB63-RNAiTC. The expression of the FaMYB63 gene was transiently upregulated or downregulated in the fruit receptacles by the OVX and RNAi approaches (Figure 3, B and C).

Figure 3

Transcript analysis of eugenol biosynthetic genes and eugenol content in transient FaMYB63-OVX and FaMYB63-silenced fruit receptacles. A, Analysis by RT-qPCR of transcript levels of genes involved in the eugenol biosynthesis pathway in FaMYB63-silenced fruit receptacles agro-infiltrated with the pHellsgate2-FaMYB63 construct, compared with control fruit receptacles agro-infiltrated with an empty vector (pHellsgate2). B, FaMYB63 silencing effect on eugenol content. FaMYB63-RNAiTA, FaMYB63-RNAiTB, and FaMYB63-RNAiTC represent three independent fruit receptacles (considered as three biological replicates) transiently injected with Agrobacterium containing FaMYB63-RNAi construct. RT-qPCR analysis of (C) FaMYB63, (D) FaEGS1, (E) FaEGS2, (F) FaCAD1 transcript levels and (G) eugenol content in strawberry fruit receptacles collected 5 d after infiltration with Agrobacterium carrying pCXSN-FLAG-FaMYB63 construct or an empty vector (pCXSN-FLAG) as a control. FaMYB63-OVX1, FaMYB63-OVX2, and FaMYB63-OVX3 represent three independent fruit receptacles (three biological replicates) transiently overexpressing FaMYB63, respectively. Error bars indicate the standard deviation of three technical replicates. Statistical significance with respect to the control (empty vector) was determined by the Student’s t test: *P < 0.05, **P < 0.01, ***P < 0.001.

FaMYB63-silenced strawberry fruits contained both significantly lower levels of transcripts related to eugenol (especially the genes EGS1, EGS2, and CAD1) and significantly lower eugenol content compared with the control group (empty vector; Figure 3, A and B). Fruits overexpressing FaMYB63 showed increases in FaMYB63 transcripts, by 63, 1,022, and 421 times in each of the three independent transformants (Figure 3C). The amounts of eugenol were significantly increased in the fruits overexpressing FaMYB63 compared with the control, corresponding to the increased transcript levels of FaEGS1, FaEGS2, and FaCAD1 (Figure 3, C–G). The transcript levels of other eugenol-related genes, namely FaC4H, FaCCR, Fa4CL, and FaPAL, did not show consistent changes in the three OVX-FaMYB63 fruits (Supplemental Figure S4). These results indicated that FaMYB63 was involved in eugenol accumulation.

FaMYB63 activates the promoters of the FaEGS1, FaEGS2, and FaCAD1 genes, which encode key enzymes of eugenol metabolism

In Arabidopsis, AtMYB63 is involved in the regulation of lignin biosynthesis, which is a branch of the phenylpropanoid biosynthetic pathway (Zhou et al., 2009). The TF FaEOBII can activate the promoters of the FaCAD1 and FaEGS2 genes (Medina-Puche et al., 2015). Previous studies reported that MYB proteins can regulate downstream genes by recognizing and binding to the MYB cis-element of a DNA sequence (Hu et al., 2017). Therefore, we hypothesized that FaMYB63 functioned by a similar mechanism. To explore the direct in vitro binding of FaMYB63 to the promoters of FaEGS1, FaEGS2, and FaCAD1, an electrophoretic mobility shift assay (EMSA) was conducted using purified recombinant FaMYB63–glutathione S-transferase (GST) fusion protein.

The promoters of FaEGS1 (1,282 bp), FaEGS2 (1,424 bp), and FaCAD1 (2,204 bp) from ‘Benihoppe’ strawberry were amplified using specific primers (Supplemental Table S1). MYB binding sites (CCAACC) between positions −251 and −300 bp of the FaEGS1 promoter (1,282 bp), (CGGTTG) between positions −1,374 and −1,424 bp of the FaEGS2 promoter (1,424 bp), and (CTGTTA) between positions −1,772 and −1,722 bp of the FaCAD1 promoter (2,204 bp) were identified (Figure 4, A, C, and E). The recombinant FaMYB63 protein could bind directly to the fragments of the FaEGS1, FaEGS2, and FaCAD1 promoters containing the MYB binding sequence (Figure 4, A, C, and E). Increasing the concentration of the unlabeled (cold) probe reduced detectable binding, while cold mutant probes failed to compete for FaMYB63 binding. FaMYB63 also displayed strong interactions with the promoters of the FaEGS1, FaEGS2, and FaCAD1 genes in Y1H assays (Figure 4, B, D, and F).

Figure 4

FaMYB63 directly interacts with the FaEGS1, FaEGS2, and FaCAD1 promoters. EMSA between FaMYB63 and the promoters of FaEGS1 (A), FaEGS2 (C), and FaCAD1 (E). Mutated nucleotides indicated with red letters. Purified GST-tagged FaMYB63 protein was incubated with unlabeled probe (cold) or biotin-labeled probe, and DNA–protein complexes were separated on native polyacrylamide gels, then photographed. The presence or absence of specific probes is marked by the symbol “+” or “−.” Specific binding of FaMYB63 to promoters of FaEGS1 (B), FaEGS2 (D), and FaCAD1 (F) in Y1H system with bait and either prey or negative control (bait/pGADT7) constructs simultaneously co-transformed into competent cells using selective growth medium (SD/-Leu) without (left) or with (right) AbA.

FaMYB63, was overexpressed in the presence of target-promoter reporter constructs using GUS assays and dual-luciferase (LUC) assays in N. benthamiana leaves. GUS staining showed that N. benthamiana leaves containing pFaEGS1:GUS, pFaEGS2:GUS, and pFaCAD1:GUS promoter constructs co-expressed with 35S:FaMYB63 exhibited much higher GUS activities than those harboring pFaEGS1:GUS, pFaEGS2:GUS, and pFaCAD1:GUS alone (Figure 5A). In addition, when either P_EGS1:LUC, P_EGS2:LUC, or P_CAD1:LUC was co-expressed with pGreenII 62-SK-FaMYB63, the LUC/REN ratios were strongly increased compared with those in cells co-transfected with CaMV35S:empty (Figure 5B). These results revealed that FaMYB63 activated GUS transcription when driven by the FaEGS1, FaEGS2, or FaCAD1 promoters. These findings demonstrated that FaMYB63 positively regulates the transcript levels of the three key genes related to eugenol production.

Figure 5

FaMYB63 regulates the transcription of FaEGS1, FaEGS2, and FaCAD1. A, Schematic diagrams of the effector (35S:FaMYB63) and reporter vectors (FaEGS1, FaEGS2, and FaCAD1 promoters fused to GUS) that were used for transient expression analysis, GUS staining, and GUS activity assays in transgenic N. benthamiana leaves. pFaEGS1:GUS, pFaEGS2:GUS, or pFaCAD1:GUS with or without the 35S:FaMYB63 effector were co-transformed into N. benthamiana leaves. The blue color indicates activated GUS activity. Data are means ± sd of nine biological replicates. B, Reporter and effector constructs used in the dual-LUC assay. LUC activity analysis, using 35S:FaMYB63 as the effector and P_EGS1:LUC, P_EGS2:LUC, or P_CAD1:LUC as reporters. Promoter activities of FaEGS1, FaEGS2, and FaCAD1 genes activated by FaMYB63 in dual-LUC assays were expressed as the ratio of LUC to REN in N. benthamiana leaves co-transformed with the effector and the reporter combinations. Data are means ± sd of nine biological replicates. Statistical significance was determined using Student’s t test: ***P < 0.001.

FaMYB63 controls FaEOBІІ and FaMYB10 transcript levels

In petunia, the hierarchical interrelationship between three R2R3-MYB TFs, EOBI, EOBII, and ODO1, has been delineated in the volatile phenylpropanoids biosynthetic pathway (Spitzer-Rimon et al., 2012). Recently, the R2R3-MYB regulator FaEOBІІ was identified and found to affect metabolic flow toward phenylpropanoid production. In addition, FaMYB10 can transcriptionally activate FaEOBІІ expression, but FaEOBІІ cannot activate FaMYB10 (Medina-Puche et al., 2014, 2015). In this study, OVX of FaMYB63 resulted in dramatic, concomitant increases in FaEOBII and FaMYB10 transcript levels and increases in both anthocyanin and eugenol contents in strawberry fruit (Figures 3, G, 6, A and B; Supplemental Figure S5A). Significant decreases of FaEOBII and FaMYB10 transcript levels were induced by FaMYB63 silencing, along with reductions in anthocyanin and eugenol contents in strawberry fruit (Figures 3, B, 6, C and D; Supplemental Figure S5B).

Figure 6

RT-qPCR analysis of FaEOBІІ and FaMYB10 transcript levels in strawberry FaMYB63-overexpressing and FaMYB63-silenced fruit receptacles (F. × ananassa cv. ‘Benihoppe’). A and B, FaMYB63-OVX1, FaMYB63-OVX2, and FaMYB63-OVX3 represent three independent FaMYB63 OVX fruit receptacles (three biological replicates), respectively. C and D, FaMYB63-RNAiTA, FaMYB63-RNAiTB, and FaMYB63-RNAiTC represent three independent fruit receptacles (three biological replicates) transiently injected with Agrobacterium containing FaMYB63-RNAi construct. All samples were collected 5 d after transient transformation. The transcript levels of both genes studied in FaMYB63-silenced fruit receptacles are expressed as a percentage against their expression levels in fruit receptacles injected with Agrobacterium containing empty vector. Error bars indicate the standard deviation of three technical replicates. Statistical significance was determined using Student’s t test: *P < 0.05, **P < 0.01, ***P < 0.001.

To investigate whether hierarchical relationships exist among FaMYB63, FaEOBІІ, and FaMYB10, we used EMSA to explore if any of the TFs are capable of physically interacting with the promoters of the other genes. The EMSA results demonstrated that the recombinant FaMYB63 protein could bind directly to the MBS-containing fragments of the FaEOBІІ and FaMYB10 promoters. However, mutations to the “CNGTTR” promoter element prevented this binding (Figure 7, A and C). In the Y1H assay, FaMYB63 displayed strong interactions with the promoters of the FaEOBІІ and FaMYB10 genes in yeast cells (Figure 7, B and D). GUS and dual-LUC reporter assays further validated these results (Figure 8).

Figure 7

FaMYB63 binds to the FaEOBІІ and FaMYB10 promoters. EMSA analysis of the binding of recombinant FaMYB63 protein to the promoters of FaEOBІІ (A) and FaMYB10 (C). The presence or absence of specific probes is marked by the symbol “+” or “−”. Specific binding of FaMYB63 to promoters of FaEOBІІ (B) and FaMYB10 (D) by one-hybrid system using bait and either prey or negative control (bait/pGADT7) constructs co-transformed into yeast competent cells using selective growth medium (SD/-Leu) without (left) or with (right) AbA.

Figure 8

FaMYB63 regulates the transcription of FaMYB10 and FaEOBІІ. A, Schematic diagrams of effector (35S:FaMYB63, 35S:FaMYB10, and 35S:FaFaEOBІІ) and reporter vectors (promoters of FaMYB63, FaMYB10, or FaEOBІІ:GUS) that were used for transient expression. GUS activity in transgenic N. benthamiana leaves infiltrated with labeled constructs. Data are means ± sd of nine biological replicates (right side). B, Schematic diagram of the reporter and effector constructs used in the dual-LUC assay. Promoter activities were expressed as the ratio of LUC to REN in N. benthamiana leaves co-transformed with effector and the reporter combinations as listed. Data are means ± sd of nine biological replicates. Statistical significance was determined using Student’s t test: ***P < 0.001.

Histochemical staining showed that GUS activity increased 3.93- and 2.62-fold in N. benthamiana leaves co-transformed with 35S:FaMYB63 and either pFaMYB10:GUS or pFaEOBІІ:GUS compared with the leaves transformed with the empty vector. There were no changes when the recombinant plasmid pFaMYB63:GUS was combined with 35S:FaEOBІІ or 35S:FaMYB10 compared with the empty vector (Figure 8A). In transient dual-LUC assays, the LUC/REN ratio was significantly increased when either the P_EOBІІ:LUC or the P_MYB10:LUC reporter constructs were co-transfected with the CaMV35S:FaMYB63 effector construct compared with when they were co-transfected with CaMV35S:empty. The LUC/REN ratio was not significantly changed when the P_MYB63:LUC reporter construct was co-transfected with the CaMV35S:FaEOBІІ or CaMV35S:FaMYB10 effectors compared with when it was co-transfected with CaMV35S:empty (Figure 8B). These findings demonstrated that FaMYB63 could regulate FaEOBІІ and FaMYB10 transcript levels, but not the other way around.

Silencing FaMYB63 in stable transgenic strawberry decreases eugenol production

To confirm the roles of FaMYB63 in eugenol metabolism in strawberry fruits, diploid strawberry (F. vesca) was stably transformed with a pHellsgate2-FaMYB63 RNAi silencing construct according to Gu and Shen (2020). RT-qPCR analysis showed that transcript levels of genes in the eugenol biosynthetic pathway corresponded to the decreased eugenol accumulation (Figures 1 and 9). The silenced lines displayed dramatically lower levels of EGS1, CAD1, EGS2, EOBII, and MYB10 transcripts compared to wild-type (WT; Figure 9A; Supplemental Figure S6). Under normal growth conditions, the transgenic lines showed significantly reduced eugenol production (Figure 9B). Moreover, fruit coloration was delayed in the FaMYB63-silenced lines compared with WT (Supplemental Figure S7). Both contents of main anthocyanins including cyanidin-3-glucoside (C3G) and pelargonidin-3-glucoside (Pg3G) and the expression of key genes involved in anthocyanin biosynthesis decreased significantly in FaMYB63-silenced lines compared with that of WT in the same developmental stages (Supplemental Figures S8 and S9).

Figure 9

Analysis of the expression of eugenol-related genes and eugenol contents in stable FaMYB63 transgenic strawberry fruits. RT-qPCR analysis of MYB63, EGS1, CAD1, EGS2, EOBII, and MYB10 transcript levels (A) and GC–MS quantification of eugenol (B) in stable FaMYB63 transgenic strawberry plants. WT, G1 stage fruits of diploid wild strawberry (F. vesca) ‘Ruegen’; FaMYB63-RNAiS1, FaMYB63-RNAiS2, and FaMYB63-RNAiS3 represent G1 stage fruits from three independent FaMYB63-silenced transgenic strawberry lines, respectively. The expression levels of the different genes studied in stable FaMYB63 transgenic strawberry ‘Ruegen’ were expressed as a percentage against their expression levels in WT strawberry plants. Error bars indicate the standard deviation of three biological replicates for each line. Statistical significance was determined using Student’s t test. ***P < 0.001.

Discussion

Previous studies identified several R2R3-type MYB TFs as positive regulators of the benzenoid/phenylpropanoid pathway in both petunia flowers and strawberry fruits (Spitzer-Rimon et al., 2012; Medina-Puche et al., 2014, 2015). In strawberry, the MYB TFs FaMYB1, FaMYB10, and FaEOBІІ have been shown to be directly connected with phenylpropanoid metabolism (Aharoni et al., 2001; Medina-Puche et al., 2014, 2015). Moreover, the expression of FaEOBII is under the control of FaMYB10 during the fruit ripening process (Medina-Puche et al., 2015). In this study, the R2R3-MYB TF FaMYB63 was identified through Y1H screening with the FaEGS1 promoter. We demonstrated that FaMYB63 functions in positively regulating eugenol production by directly activating the expression of FaEGS1, FaEGS2, and FaCAD1 during fruit development. FaMYB63 was also shown to activate the transcription of FaEOBІІ and FaMYB10, and thereby enhance eugenol accumulation.

Tissue specificity and developmental regulation of FaMYB63

As postulated by Fait et al. (2008), the metabolic program of the achene is largely concerned with accumulating storage and protective compounds as well as precursors for hormonal and secondary metabolites. Eugenol is produced by plants as a defense compound against animals and microorganisms, as a floral attractant of pollinators, and as a key volatile phenylpropanoid compound in fruits to attract consumers for seed dispersal (Koeduka et al., 2006; Aragüez et al., 2013). Our results indicated that the transcript levels of FaMYB63 were relatively higher in the achenes than in the receptacles, leading to more eugenol accumulation in the achenes.

Strawberry is an aggregate fruit that originates from the swollen receptacle with a few hundred achenes embedded in the epidermis of the receptacle. The achenes are the true fruit, while the receptacle is an accessory fruit resulting from cell enlargement and cell division. Both the achene and the receptacle constitute the fleshy part (Fait et al., 2008). Gene expression patterns in achenes were distinct from that in the receptacles during fruit development and maturation (Aharoni and O’Connell, 2002). Our results of FaEGS1 and 2 expression patterns (Supplemental Figure S10) were consistent with the above-mentioned previous study (Molina-Hidalgo et al., 2017) showing FaEGS1 was mostly expressed in the receptacles at the green stage, whereas FaEGS2 expression is majorly restricted to red receptacles. FaCAD1 showed clear correlation with the expression profile of FaMYB63 as well as eugenol content in receptacles. In addition, FaMYB63 is expressed in both achenes and receptacles and can activate FaEGS1, FaEGS2, and FaCAD1, suggesting that FaMYB63 might regulate the eugenol accumulation in both tissues, although its expression is higher in the achenes than in the receptacles. The ripening-related TF genes, for example, FaEOBII and FaMYB10, were also expressed in both tissues, but their transcription levels were higher in receptacles than in achenes (Supplemental Figure S10).

Our study found that the expression pattern of FaMYB63 was largely different from those of other ripening-related and receptacle-specific R2R3 regulators involved in eugenol biosynthesis in strawberry fruits, for example, FaMYB10 and FaEOBII (Medina-Puche et al., 2014, 2015), but did correlate with eugenol content at different developmental stages. Furthermore, the developmental stage-related changes in eugenol content coincided with Medina-Puche et al. (2015). In addition, the peak expression levels of FaMYB63, FaEGS1, and FaCAD1 were found in tissues other than flowers, such as small green receptacles and achenes, where the amounts of eugenol were the highest (Figure 2, B and C; Supplemental Figure S10). The expression pattern of FaEOBII was similar to that of FaEGS2 (Supplemental Figure S10) and the expression of FaEOBII and FaEGS2 presented higher levels in petals than in red-ripe receptacles (Medina-Puche et al., 2015). Therefore, FaEOBII would be dominant in determining eugenol accumulation in petals, while FaMYB63 could be the primary regulator during early fruit development in small green receptacles and achenes. The transcript level of FaEOBII was lower than that of FaMYB63 in red-ripening fruit (Supplemental Figure S11). Collectively, FaMYB63 would be, in quantitative terms, a major player in the regulation of eugenol accumulation.

FaMYB63 expression is suppressed by ABA and promoted by auxins during fruit ripening

The balance between ABA and auxins triggers strawberry fruit ripening and induces complex physiological and metabolic effects (Perkins-Veazie, 1995; Medina-Puche et al., 2014, 2015). Auxins, secreted from achenes, have an antagonistic influence on green receptacle growth during the early fruit ripening stage (Kang et al., 2013), while ABA plays pivotal roles in the regulation of strawberry fruit ripening at the late stage and in adaptation to abiotic environmental stresses (Finkelstein et al., 1998; Jia et al., 2011). Ripening-related processes that are sensitive to auxin include softening, color change, and aroma formation (Aharoni and O’Connell, 2002). Our results indicated that the TF FaMYB63 expression was regulated by auxins and ABA in the opposite way from most ripening genes.

The vast majority of genes involved in strawberry ripening are co-regulated by auxin and ABA, for example, FaQR (Raab et al., 2006; Daminato et al., 2013), FaPG1 (Quesada et al., 2009; Daminato et al., 2013), FaRGLyase1 (Molina-Hidalgo et al., 2013), FaNIP1;1 (Molina-Hidalgo et al., 2015), and FaβGal4 (Paniagua et al., 2016), in addition to ripening-related TFs such as FaSHP (Daminato et al., 2013), FaMYB10 (Medina-Puche et al., 2014), FaEOBII (Medina-Puche et al., 2015), FaDOF2 (Molina-Hidalgo et al., 2017), and FaPRE1 (Medina-Puche et al., 2019). These molecular data show the positive effects of ABA and the negative effects of auxin on the expression of ripening-induced genes. However, the expression of FaMYB63 decreased prematurely in receptacles when achenes were removed, and the internal auxin content declined during the early stages of fruit development. The expression level of FaMYB63 largely increased when auxin (IAA) was applied to fruits from which the achene was removed, implying that this gene expression was positively regulated by auxins (Figure 2, D and E).

In this study, two experimental approaches (according to Medina-Puche et al. 2015) were used to ascertain the role of ABA in the regulation of FaMYB63 expression. Inhibition of ABA synthesis by NDGA significantly increased FaMYB63 expression and eugenol content. Moreover, in strawberry subjected to water stress, ABA content increased and FaMYB63 expression decreased (Figure 2, F–I). The expression patterns of FaMYB63 in response to ABA and auxins were similar to FaMADS1a, which might be involved in fruit ripening and play a negative role in anthocyanin accumulation of strawberry fruit (Lu et al., 2018). These dual regulation results suggested that hormonal regulation of FaMYB63 was different from the other two TFs (FaMYB10 and FaEOBII) that drive eugenol biosynthesis regulation and from many other functionally characterized genes involved in the biosynthesis of compounds related to fruit organoleptic properties (Raab et al., 2006; Quesada et al., 2009; Daminato et al., 2013; Molina-Hidalgo et al., 2013, 2015, 2017; Medina-Puche et al., 2014, 2015, 2019; Paniagua et al., 2016).

FaMYB63 regulates the expression of three genes that play prominent roles in eugenol biosynthesis in strawberry fruits

FaEGS1, FaEGS2, and FaCAD1 encode key enzymes of the metabolic pathway that produces eugenol (Blanco-Portales et al., 2002; Aragüez et al., 2013). FaEOBII directly regulates the expression of both the FaCAD1 and FaEGS2 genes and affects eugenol biosynthesis in strawberry (Medina-Puche et al., 2015). Previous studies in petunia showed that EOBI silencing led to downregulation of ODO1 and of numerous genes from phenylpropanoid pathways that synthesize scent-related compounds, for example, S-adenosyl-L-methionine:benzoic acid/salicylic acid carboxyl methyltransferase 1 and 2 (BSMT1,2), isoeugenol synthase (IGS), and EGS (Spitzer-Rimon et al., 2012).

In this study, FaMYB63-silenced strawberry fruits showed downregulation of many genes involved in the downstream benzenoid/phenylpropanoid pathway, suggesting that some biosynthetic genes may be regulated by FaMYB63. EMSA and Y1H assays demonstrated that FaMYB63 was able to bind to the MBS boxes in the promoters of FaEGS1, FaEGS2, and FaCAD1 (Figure 4). GUS assays and dual-LUC assays confirmed that FaMYB63 could also bind directly to the three biosynthetic genes and significantly enhance their activities (Figure 5). Silencing in the transient expression assay and in stable transgenic plants, decreased the transcript levels of the three biosynthetic genes and the eugenol content (Figures 3 and 9). Our findings also showed that overexpressing FaMYB63 increased FaEGS1, FaEGS2, and FaCAD1 transcript levels, leading to an increased eugenol content in the fruits (Figure 3, C–G).

R2R3-MYB TFs, acting as activators or repressors, are involved in divergent physiological and biochemical processes, including developmental control and cell fate determination, response to environmental stimuli, and secondary metabolism (phenylpropanoid metabolism and its branches, such as anthocyanins and flavonoids) (Weisshaar and Jenkins, 1998; Jin and Martin, 1999). For example, VvMYB14 is involved in the melatonin signaling pathway and promotes ethylene production by transcriptionally activating 1-aminocyclopropane-1-carboxylate synthase 1 (VvACS1), thereby altering the accumulation of secondary metabolites in grapes (Vitis vinifera; Ma et al., 2021). VvMYB14 is also reported to transcriptionally regulate stilbene biosynthesis by specifically inducing stilbene synthase (STS) transcription (Höll et al., 2013). In strawberry, FaMYB1 acts as a repressor and suppresses anthocyanin and flavonol accumulation in transgenic tobacco (Nicotiana tabacum; Aharoni et al., 2001). MYB10 is the essential positive activator of the anthocyanin and flavonoid pathways (Medina-Puche et al., 2014; Castillejo et al., 2020). In ripe strawberry fruits, FaMYB10 could reverse FaMYB44.2 repression of sucrose-6-phosphate synthase 3 (FaSPS3), resulting in sucrose accumulation (Wei et al., 2018). In our study, FaMYB63 acted as a positive regulator of eugenol production by directly binding to the promoters of FaEGS1, FaEGS2, and FaCAD1. In contrast, FaEOBII increases eugenol accumulation by triggering the expression of only two downstream target genes, FaCAD1 and FaEGS2 (Medina-Puche et al., 2015).

FaEGS1 was highly expressed in green fruit, which was similar to the FaODO1 and FaDOF1 expression patterns, whereas the FaEGS2 transcript was highly expressed in ripe fruit (Aragüez et al., 2013; Molina-Hidalgo et al., 2017). However, at early and later developmental stages of strawberry fruit, FaMYB63 transcription levels were higher than those of FaEOBII (Supplemental Figure S11); therefore, FaMYB63 could regulate FaEGS1, FaEGS2, and FaCAD1. In many different developmental regulatory contexts, temporal and spatial interactions of gene expression are not mediated solely by one regulator (Aharoni et al., 2001). Prior studies have reported that at least six R2R3-MYB superfamily TFs play important roles in glucosinolate biosynthesis in Arabidopsis (Zhang et al., 2018). In fruit where FaMYB63 gene expression was silenced or enhanced, concomitant eugenol production was not highly reduced or increased, similarly to the functions of FaEOBII in a previous report (Medina-Puche et al., 2015). FaODO1 and FaDOF1 TFs may regulate the expression of FaEGS1 in green unripe fruits (and thus eugenol biosynthesis); moreover, FaEOBII activates the transcription levels of FaEGS2 in coordination with FaDOF2 (Molina-Hidalgo et al., 2017). FaMYB10, at higher levels in the red-ripe fruit stage than in the small green fruit stage, would activate the expression of FaEOBII, which would subsequently induce the expression of FaCAD1 and FaEGS2 and eugenol production in ripe strawberry (Medina-Puche et al., 2015). Taken together, these studies show that the regulatory network responsible for eugenol production in strawberry fruit is multifaceted.

FaMYB63 expression is not regulated by FaMYB10 or FaEOBII

Complex transcriptional regulation of metabolic networks has been detailed in some plants. In petunia, the emission of phenylpropanoid volatiles was regulated by a complex network of interactions between R2R3-MYB-like regulatory factors. EOBII directly regulates EOBI expression, and both EOBI and EOBII activate the transcript level of ODO1, a distant relative of EOBI and EOBII, while ODO1 in turn negatively influences the expression of EOBI (Spitzer-Rimon et al., 2012). In Arabidopsis, MYB99 (a putative ortholog of petunia ODO1), MYB21, and MYB24 (orthologs of EOBI and EOBII) act as an analogous regulatory module controlling phenylpropanoid metabolism (Battat et al., 2019).

In strawberry, FaMYB10 has a broad regulatory effect, including on 14 genes encoding early and late biosynthetic enzymes of flavonol/phenylpropanoid metabolism and multi-drug and toxic compound extrusion or GST transporters (Medina-Puche et al., 2014). FaMYB10 activates the expression of FaEOBII to induce eugenol production (Medina-Puche et al., 2015). Several TFs have been shown to regulate anthocyanin biosynthesis via interaction with MYB10. SCARECROW-LIKE 8 can modulate the transcriptional regulation of genes related to flavonoid/anthocyanin biosynthesis, probably through its influence on FaMYB10 expression (Pillet et al., 2015). FaMYB10 can be bound by FaRAV1 to stimulate anthocyanin accumulation (Zhang et al., 2020). In apple (Malus × domestica), MdHY5 promotes anthocyanin production by directly binding to the MdMYB10 promoter (An et al., 2017). In this work, silencing and overexpressing FaMYB63 decreased and increased, respectively, the transcript levels of FaEOBII and FaMYB10 in strawberry fruits. Moreover, there were concomitant reductions and increases in anthocyanin and eugenol contents (Figures 3 and 6; Supplemental Figure S5). Our results suggested that FaMYB63 is also involved in anthocyanin biosynthesis, possibly by regulating FaMYB10 (Supplemental Figures S7, S8, and S9). Of course, the mechanisms of how FaMYB63 regulates anthocyanin accumulation need further investigations. FaEOBII could subsequently trigger FaCAD1 and FaEGS2 gene expression (Medina-Puche et al., 2015), while FaMYB63 directly activated FaEGS1, FaEGS2, and FaCAD1 to induce eugenol production. However, FaMYB63 expression was not regulated by FaEOBII or FaMYB10 (Figures 1 and 8). The regulatory function of FaMYB63 is similar to that of PhEOBII in petunia petals. In petunia, EOBII directly activates EOBI and ODO1 and transactivates at least two genes encoding scent-related phenylpropanoid biosynthetic enzymes, IGS and PAL, but the expression of EOBII is not modulated by EOBI and ODO1 (Van Moerkercke et al., 2011; Spitzer-Rimon et al., 2012).

Conclusion

Previous studies reported that the R2R3-MYB TFs involved in eugenol accumulation mainly belong to subgroup 19, such as PhEOBI, PhEOBII, and FaEOBII. In this work, another R2R3-type MYB TF, FaMYB63, distant from the TFs mentioned above, was identified. This TF enhanced eugenol production as a positive regulator by activating transcription of not only the regulatory genes FaEOBІІ and FaMYB10, but also the genes encoding the key eugenol biosynthetic enzymes FaEGS1, FaEGS2, and FaCAD1 in the early and late stages of strawberry fruit development. The transient OVX, RNAi, and stable transformation indicated that FaMYB63 might be also involved in anthocyanin production. This study provides insight into the regulatory networks in the eugenol synthesis pathway in strawberry.

Materials and methods

Plant materials and growth conditions

Strawberry (an octoploid cultivar, F. × ananassa ‘Benihoppe’; a diploid wild strawberry, F. vesca ‘Ruegen’) plants were grown in a greenhouse in the Strawberry Germplasm Resource Garden of Anhui Agricultural University with a diurnal rhythm of 16-h light and 8-h darkness following normal cultivation practices. The greenhouse environment was maintained at 25°C in the day and 18°C in the night with 60% humidity. Seedlings were watered daily to the drip point. Fruit samples were collected from four to five individual plants beginning with the small green fruit stage (10-d postanthesis [DPA]). Between 10 and 18 fruits of the same maturation degree were harvested every 5 d, with sample collection ending at the OR stage (5 d after the red-ripening stage). These fruit samples coincided with the fruit ripening stages of small green fruit (Stage 1, G1), large green fruit (Stage 2, G2), green-white fruit (Stage 3, G3), white fruit (Stage 4, W), red-turning fruit (Stage 5, T), red-ripening fruit (Stage 6, R), and over-ripening fruit (Stage 7, OR). Stolons, roots, flowers, and expanding leaves were also harvested. All fruits and tissues from different stages were washed in water, cut into quarters, pooled, immediately frozen in liquid nitrogen, and stored at −80°C. Nicotiana benthamiana plants used for dual-LUC and GUS assays in this study were grown in a growth chamber with a light/dark cycle of 16-h/8-h at 24°C.

RNA isolation and RT-qPCR

RNA extraction and transcript analysis were carried out as described by Xie et al. (2012). Three biological replicates were tested for all samples using the interspacer 26S-18S RNA gene from strawberry as an internal control (Medina-Puche et al., 2015). The primers used are listed in Supplemental Table S2. The relative expression levels were calculated using the 2^–△△Ct method (Livak and Schmittgen, 2001).

Cloning full-length cDNA of FaMYB63 and sequence analysis

A partial MYB sequence was obtained from a cDNA library of strawberry fruit using a Matchmaker Gold Y1H library screening system (Takara, Kyoto, Japan). The sequence was used as a query in a BLAST search against Genome Database for Rosaceae (GDR) (https://www.rosaceae.org/) and National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov) databases. The gene with the best matches to the partial sequence was considered a candidate gene, and primers (Supplemental Table S1) were designed to amplify the full-length coding sequence (CDS) from cDNA isolated from ‘Benihoppe’ fruits. Conserved domains within the predicted FaMYB63 protein were analyzed by CD-search (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi).

The MYB proteins selected for the construction of a phylogenetic tree were obtained from NCBI (Supplemental Table S3). Multiple sequence alignment was processed into a maximum-likelihood phylogenetic tree using MEGA version 7.0, with bootstrapping of 1,000 iterations for validation.

Generation of the OVX-FaMYB63 and RNAi-FaMYB63 construct and transfection of strawberry fruits by agroinfiltration

The full-length CDS of FaMYB63 from fruit was isolated by RT-PCR from strawberry fruit cDNA using primers as described in Supplemental Table S1. These products were linked into the pMD19-T vector and were subsequently transformed into Escherichia coli DH5α. Positive colonies were selected, amplified, and then sequenced by Tsingke Biotechnology Co., Ltd. (Nanjing, China). To generate the 35S_pro:FaMYB63 construct, the full-length cDNA sequences were cloned into the binary expression vector pCXSN-FLAG (Chen et al., 2009). To generate the FaMYB63-RNAi construct, fragments of the ORF of FaMYB63 (769–1,105 bp) were PCR-amplified and fused to generate a 336-bp PCR product containing the attB1 and attB2 sites, and the fused PCR product was cloned into pHellsgate2 (Invitrogen, Waltham, MA, USA; Wesley et al., 2001). Empty vectors, pHellsgate2-FaMYB63 or pCXSN-FLAG-FaMYB63 constructs were transformed into A. tumefaciens strain GV3101 by the freeze–thaw method.

Agrobacterium tumefaciens containing pCXSN-FLAG-FaMYB63 or pHellsgate2-FaMYB63 was used to inject strawberry fruits (F. × ananassa cv. ‘Benihoppe’), according to Medina-Puche et al. (2015). OVX or interference of FaMYB63 was carried out by co-infiltration of A. tumefaciens harboring the silencing suppressor pSoup-p19 (Havelda et al., 2003). The A. tumefaciens strain harbouring empty vector or recombinant plasmids was grown at 28°C in liquid Luria Bertani (LB) medium with appropriate antibiotics. The cells were collected by centrifugation and resuspended in infiltration buffer (10-mM MgSO₄, 10-mM MES, and 1-mM acetosyringone). The transformed cells carrying construct or empty vector were left at room temperature for 2–4 h in darkness. Agrobacterium tumefaciens culture (0.2 mL, optical density at 600 nm (OD₆₀₀) was 0.6) containing individual RNAi or OVX constructs or the empty vector was injected into the strawberry fruits using a 1-mL sterile syringe. For each vector, at least seven fruits were agro-injected. The data for “Empty vector” were mean ± sd obtained from all the injected fruits (n ≥ 7, considered as seven biological replicates). For FaMYB63-OVX (or FaMYB63-RNAi) vector, the successfully injected fruits were screened out by using RT-qPCR. The top three fruits showing the highest (or lowest) expression of FaMYB63 were named FaMYB63-OVX1, –OVX2, and -OVX3 (or FaMYB63-RNAiTA, TB, and TC), respectively, and used for further analysis. The three selected fruits were considered as three biological replicates. The data for each selected fruit were mean ± sd obtained from three technical replicates. In all cases, the RNAi percentages were determined by comparing the amounts of FaMYB63, FaEGS1, FaEGS2, FaCAD1, FaEOBII, and FaMYB10 transcripts in fruits agro-injected with the RNAi construct against those observed in fruits agro-injected with empty pHellsgate2 vector (Medina-Puche et al., 2015).

Subcellular localization experiments

The CDS of FaMYB63 was amplified by PCR using forward primers with a BglII-restriction site (5′-CGAGATCTCATGGTAGGGAGGGGAAGAAC-3′) and reverse primers with a SpeI-restriction site (5′-CGACTAGTGTTAGCGGAGTTCTCTGGCA-3′) and subcloned into the expression vector pCAMBIA1302 under the control of the CaMV35S promoter. The generated GFP construct (pCAMBIA1302-FaMYB63) was sequenced. Transient expression of FaMYB63-GFP translational fusions was carried out as described by Xing et al. (2019).

Agroinfiltration experiments were performed on N. benthamiana leaves. The Agrobacterium strain GV3101 (carrying pSoup-p19) containing either pCAMBIA1302-FaMYB63 or empty pCAMBIA1302 was grown at 28°C in LB medium with appropriate antibiotics. Cells were harvested and resuspended in infiltration buffer (10-mM MgSO₄, 10-mM MES, and 1-mM acetosyringone). This step was repeated once, and the concentration of the bacterial suspension was measured by spectroscopy (OD₆₀₀) and adjusted to a final concentration of 0.5–0.8. The transformed cells carrying either construct were left at room temperature for 2–4 h in darkness. Bacterial suspensions were syringe-infiltrated through the abaxial surface of the leaves, and material was collected after 3 d. To locate nuclei, N. benthamiana leaf tissues were incubated with DAPI. Leaves were observed by an FV1000 confocal laser scanning epifluorescence microscope (Olympus Corporation, Tokyo, Japan) with a 40 × objective. Detector gains were set at 400 nm for GFP and GFP fluorescence was detected with setting excitation wavelength at 488 nm (10% intensity) and emission wavelength set at 500–515 nm. DAPI signal was determined using an excitation wavelength of 405 nm and an emission wavelength range of 449–461 nm.

Stable transgenic silencing of FaMYB63 in strawberry

Stable transgenic diploid strawberry was established as previously described (Gu and Shen, 2020).

Solid-phase microextraction of eugenol

The extraction of eugenol and gas chromatograhy-mass spectrometry (GC-MS) analysis were carried out as reported (Zorrilla-Fontanesi et al., 2012; Medina-Puche et al., 2015), with slight modification. Eugenol was analyzed in ‘Benihoppe’ fruits infiltrated with Agrobacterium carrying pCXSN-FLAG, pCXSN-FLAG-FaMYB63, pHellsgate2, or pHellsgate2-FaMYB63 at different developmental stages and in different tissues. Fruits were ground to a fine powder under liquid nitrogen. Up to 5 mL of saturated NaCl was added to 0.5 g of the powdered fruits in a 25 mL vial. Solid-phase microextraction was performed on a 7890B GC system (Agilent Instruments, Santa Clara, CA, USA) coupled to a 7000 GC–MS Triple Qual quadrupole mass detector (Agilent). Samples were incubated for 15 min at 50°C, and then volatiles were extracted on a 50/30 μm DVB/CAR/PDMS fibre (Supelco, Bellefonte, PA, USA) for 30 min at 50°C.

The extracts were desorbed for 1 min at 250°C in the injector port and then injected in splitless mode. The volatiles were separated on a DB-5 MS column (0.25 mm × 60 m, 0.25 μm, DB-5 MS stationary phase, Agilent 122-5562UI, CA, USA) with a constant helium flow of 1.0 mL min⁻¹. The oven temperature was set to 40°C for 5 min, ramped at 5°C min⁻¹ up to 250°C, and held at 250°C for 5 min. The significant MS operating parameters were as follows: the ion source temperature was 230°C and the interface temperature was 250°C. Twenty mass spectra were recorded every second (20–400 amu) at an ionization energy of 70 eV. Ten strawberry fruits were harvested per sample, and three samples were measured on each occasion. The eugenol was quantified on the basis of a standard curve using the eugenol reference standard (Sigma-Aldrich, St Louis, MO, USA).

Auxin (IAA), water stress, NDGA treatments, and HPLC analysis of ABA and IAA

Auxin, water stress, and NDGA treatments were conducted according to previous reports (Medina-Puche et al., 2014, 2015). ABA and IAA analyses were performed according to the method proposed by Guinn et al. (1986).

Anthocyanidin quantification

Total anthocyanidin was extracted and determined according to Xie et al. (2012). The main anthocyanin (C3G and Pg3G) contents were determined using high-performance liquid chromatography (HPLC) as previously described method (Gao et al., 2020). Briefly, 0.5 g of strawberry fruits were ground into a powder with liquid nitrogen, 1-mL extraction solution (methanol: H₂O: hydrochloric acid, 80:19:1) was added and extracted in the dark at 4°C for 16 h. The mixture was centrifuged at 6,000 g for 10 min. The supernatant (0.5 mL) was then filtered through a 0.22 μm organic phase filter into a 2-mL brown sample vial for further analysis. The HPLC analysis was performed on an Agilent 1260 HPLC system (Agilent, Böblingen, Germany) and chromatographic separation was performed with an XSelect HSS T3 C18 column (4.6 × 250 mm, 5 μm; Waters Co., Milford, MA, USA). The mobile phase consisted of 0.1% (v/v) aqueous formic acid (solvent A) and methanol (solvent B), run at a flow rate of 1.00 mL min⁻¹. The linear gradient of phase B was as follows: 0–10 min, 5%–25%; 10–20 min, 25%–40%; 20–50 min, 40%–65%; 50–70 min, 65%–10%; and 70–75 min, 10%–5%. Samples were detected using a UV-visible light detector, and the detection wavelength was 530 nm. C3G and Pg3G were used as standards.

Y1H assay

The Y1H screening was performed using the Matchmaker Gold Y1H Library Screening System (Takara, Kyoto, Japan) and Yeastmaker Yeast Transformation System 2 (Takara) according to the manufacturer’s instructions. The amplified 1,282-bp fragment of the FaEGS1 promoter was cloned into pAbAi vector to construct the bait, and positive clones were identified by DNA sequencing.

An Y1H assay was also conducted to examine the ability of FaMYB63 to bind to the full-length promoters of FaEGS1 (1,282 bp), FaEGS2 (1,424 bp), FaCAD1 (2,204 bp), FaEOBII (2,000 bp), and FaMYB10 (1,635 bp). The promoters of the five genes were PCR-amplified with specific primers (Supplemental Table S1) and cloned into the pAbAi vector to construct bait vectors. The FaMYB63 ORF (1,134 bp) was fused in frame with the GAL4 activation domain (AD) in the pGADT7-AD vector to generate the prey vector (pGADT7-FaMYB63). The prey vector was transformed into the bait-reporter strain. Yeast cells co-transformed with the prey and bait vectors were plated on synthetic dropout (SD)/-Leu medium supplemented with or without Aureobasidin A (AbA) and incubated for 2–5 d at 30°C.

EMSA

The full-length CDS of FaMYB63 was amplified and inserted into the pGEX4T-1 vector to generate the recombinant GST-FaMYB63 plasmid. The recombinant plasmid was introduced into BL21 (E. coli) cells, and GST–FaMYB63 fusion proteins were expressed in E. coli cells essentially as described in a previously published study (Van Moerkercke et al., 2011). About 5 μg of purified recombinant GST-FaMYB63 was used for each assay.

Conserved cis-element motifs of the five promoters were predicted using PLACE (http://www.dna.affrc.go.jp/PLACE/signalscan.html) and PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). Oligonucleotide probes containing MYB-binding sites were synthesized and labeled with biotin at the 3′-hydroxyl end of the sense strand according to the manufacturer’s instructions (EMSA probe biotin labeling kit; Beyotime, Jiangsu, China; GS008). The primers used for EMSA are listed in Supplemental Table S4.

The EMSA was conducted as described in Xie et al. (2012). To confirm the specificity of the shifted band, 100-fold amount of nonlabeled identical or mutated oligonucleotides were incubated with nuclear proteins 20 min before the addition of labeled oligonucleotides. The signal was detected using LightShift Chemiluminescent EMSA Kit (Beyotime; GS009).

GUS activity assays

The putative promoter regions for FaEGS1 (1,282 bp), FaEGS2 (1,424 bp), FaCAD1 (2,204 bp), FaEOBII (2,000 bp), and FaMYB10 (1,635 bp) were amplified from the ‘Benihoppe’ genomic DNA and cloned into the pCAMBIA1391Z vector to drive the GUS reporter gene (Wang et al., 2019). The full-length cDNA of FaMYB63 was cloned into the pGreenII 62-SK vector to generate effector constructs. pFaEGS1:GUS, pFaEGS2:GUS, pFaCAD1:GUS, and 35S:FaMYB63 and pFaEOBII:GUS, pFaMYB10:GUS, and 35S:FaMYB63 or pFaMYB63:GUS and 35S:FaEOBII and 35S:FaMYB10 were independently transformed into A. tumefaciens strain GV3101 to be used for A. tumefaciens-mediated transient expression as described previously (Yang et al., 2000). The empty vector pGreenII 62-SK was used as a negative control. Six-week-old N. benthamiana leaves were infiltrated, and 3 d after infiltration the injected leaf tissues were collected. The N. benthamiana leaf discs were stained with 5-bromo-4-chloro-3-indolyl-β-d-glucuronide for 24 h at 37°C following the procedure described in Liu et al. (2012) and were then incubated in 70% ethanol to remove chlorophyll before photographing. Quantitative analysis of GUS was carried out using 4-methylumbelliferyl-β-d-glucuronide (Sigma-Aldrich, St Louis, MO, USA) as a substrate for the fluorometric assay as previously described by Jefferson (1988), and the 4-methylumbelliferone (4-MU) produced in the GUS reaction was measured using a Thermo Scientific Microplate Reader (Thermo Scientific, Waltham, MA, USA; http://www.thermofisher.com/). Final GUS activity was calculated according to the standard curve of 4-MU (Sigma-Aldrich). All analyses were repeated nine times, and the mean values for each construct were compared.

Dual-LUC activity assay

Dual-LUC assays were performed as previously described (Liu et al., 2017). The promoter fragments of FaEGS1, FaEGS2, FaCAD1, FaEOBII, and FaMYB10 were fused into pGreenII 0800-LUC to obtain five reporters. In the same vector, Renilla LUC (REN) under the control of the 35S promoter was used as a positive control. The constructed effector (62-SK-FaMYB63) and reporter plasmids were transfected into A. tumefaciens strain GV3101 separately and co-infected into N. benthamiana leaves. Three days after agroinfiltration, the ratio of LUC to REN activity was assessed using dual-LUC assay kits (Promega, Madison, WI, USA). At least nine assay measurements were included for each combination.

Statistical analysis

Statistical analysis was performed using SPSS software. Data are reported as the mean ± sd. Asterisks indicate significant differences between each treatment, assessed by Student’s t test. Different lowercase letters above the bars indicate a significant difference from each other according to Duncan’s multiple range test (P < 0.05).

Accession numbers

Sequence data from this article can be found in the GenBank/EMBL or GDR data libraries under accession numbers: FaMYB63 (MW452942); FaEOBII (KM099230); FaMYB10 (EU155162); FaEGS1 (KF562264); FaEGS2 (KF562266); FaCAD1 (U63534); FaPAL (AB360390); FaC4H (DQ898278); FaCCR (JX290510); Fa4CL (XM_004304354); FaNCED1 (HQ290318); FvCHS2 (gene26826); FvANS (gene32347); and FvDFR1 (gene15174).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Alignment of FaMYB63 to R2R3-MYB proteins.

Supplemental Figure S2. Phylogenetic analysis of FaMYB63 and its homologous proteins.

Supplemental Figure S3. Variations in transcripts of FaMYB63 and eugenol content.

Supplemental Figure S4. Expression levels of four other eugenol biosynthetic genes in control (empty vector) and FaMYB63-OVX fruit receptacles.

Supplemental Figure S5. FaMYB63 is involved in anthocyanin accumulation.

Supplemental Figure S6. FaMYB63 stable transgenic ‘Ruegen’ strawberry lines.

Supplemental Figure S7. Representative pictures of FaMYB63 stable transgenic strawberry lines at different times.

Supplemental Figure S8. Contents of main anthocyanin in control (WT) and FaMYB63-RNAiS (silenced line) fruits by HPLC at three ripening stages (14, 19, and 24 DPA).

Supplemental Figure S9. The expression of anthocyanin-related genes in control (WT) and FaMYB63-RNAiS fruits at three ripening stages (14, 19, and 24 DPA).

Supplemental Figure S10. Transcript level analysis of FaEGS1, FaEGS2, FaCAD1, FaEOBII, and FaMYB10 in fruit receptacles and achenes at different developmental stages in F. × ananassa cv. ‘Benihoppe’.

Supplemental Figure S11. Expression of FaMYB63, FaMYB10, FaEOBII, FaEGS1, FaEGS2, and FaCAD1 in octoploid strawberry (F. × ananassa cv. ‘Benihoppe’).

Supplemental Table S1. Primer sequences used for PCR.

Supplemental Table S2. Primer sequences used for RT-qPCR.

Supplemental Table S3. Subgroup classification of MYB sequences examined in this study and additional information.

Supplemental Table S4. EMSA probes.

C.B.F., J.Z., X.B.X., L.Z.Z., P.P.S., H.F., and H.X. conceived and designed the experiments. J.Z., C.B.F., and S.S.W. wrote the manuscript. S.S.W., M.Y.S., Y.Z., and Z.F.P. performed the experiments and collected and analyzed the data. All authors have read and approved the manuscript for submission.

The authors responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) are: Congbing Fang ([email protected]) and Jing Zhao ([email protected]).

Acknowledgments

The authors thank Professor Yongsheng Liu, Songhu Wang, and Peijin Li for critical reading of the manuscript and providing valuable comments.

Funding

This work was funded by the National Key R&D Program of China (2018YFD1000200), Key R&D Program of Anhui Province (201904a06020049), and the National Natural Science Foundation of China (31701869).

Conflict of interest statement: The authors declare that they have no conflicts of interest.

Author notes

†

Senior authors.